Science of the Top 40
By Erin Froehlich More Blogs by This AuthorFrom the Id and Ego Blog Series
Whether you love it or love to hate it, pop music has you down to a science – and I do mean that literally. Researchers just released an equation for predicting hit songs, backing up previous findings that showed pop music was, well….. pretty much brain crack.
Those opposed to pop music frequently site its formulaic nature as their reasoning –
Rail thin blonde with a mediocre voice+
Skimpy clothing +
Repetitive lyrics +
The right beat
INSTANT POP HIT!!
Now, it turns out they’ve have been closer to the truth we had ever thought - there really IS a pop hit formula - of course, it’s a bit more scientific than the one above. ;)
As they described on their website ScoreAHit.com, “You can see that we use the energy, tempo, danceability, loudness and other higher-level features such as harmonic simplicity (how simple the chord sequence is) and non-harmonicity (how 'noisy' the song is)….we first looked at all the UK hits for a certain time (more accurately: a computer algorithm we designed did this for us!), and measured their audio features (loudness, tempo etc). From this we got a list of weights, telling us how important each of the 23 features are.”
So rather than skimpy clothing + repetitive lyrics, we end up with a formula that looks more like:
In this equation they explain, the w’s represent the 23 different aspects of each song that they measured. The f’s on the other hand, have to do with the changing music trends of each era. They represent the “weight”, or importance of that song aspect be it loudness, danceability or whatever, to the era it was produced in.
Explained Dr. Tijl De Bie, Senior Leccturer in Artificial Intelligence, “Musical tastes evolve, which means our 'hit potential equation' needs to evolve as well. Indeed, we have found the hit potential of a song depends on the era. This may be due to the varying dominant music style, culture and environment."
Finally, the calculated score gives them a number, and based on that number, researchers can predict - with about 60% accuracy whether the song will hit the top 5 OR whether it will never make it above position 30.
“Wait …only 60%?
That seems like a lot of mistakes….”
And yes, there are a number their algorithm couldn't predict - Alicia Keys and Jay-Z's "Empire State of Mind (Part II)" released in 2010 for example, hit #4 despite it's being softer and less danceable than most hit songs of that year.
HOWEVER, when you consider the complexity of changing trends and personal taste, 60% really IS very impressive. At the very least, they can say that no one has ever made a formula that got any closer!
Applying the Formula
Now here’s where things get really interesting - when their formula was actually applied and trends began emerging. Again, quoting from their site, “Looking at the UK chart from the last 50 years, we noticed some interesting patterns. Plotting feature values through time, we can look at hits and non-hits and see how they differ. In the plots below, hits are shown in blue, non-hits in red.”
Here are some of their most interesting findings:
Simplicity Vs. Complexity
Past: While in the past, the more harmonically simple the song, the more likely it was to be a hit, by the late 80’s that trend had reversed itself.
While their researchers can’t be sure exactly why trends appear as they do, they speculate that the appearance of new generes and subgenres ahs something to do with that.
Future: Now, intriguingly, you may also notice that as we’ve hit 2010, those trends are beginning to reverse once more.
Dancing Vs. Listening Music
Past: As you can see from their chart here, the “danceability” of the song wasn’t really important to the music of the 60’s and 70’s.
It wasn’t until the early electronic hits of the 80’s that danceable beats became essential to a song’s success, but by that time as you can see from the wide space between the two lines, low danceability REALLY broke a song’s top 5 chances!
Future: Now, while danceability remains an important factor for a song’s popularity, it does seem like were on a downward trend from our peak dance hit craze in 2000.
Past: For whatever reason, song volume has been on a steady upward trend both in hit and non-hit songs since the 60’s!
Future: Music trends peaked in loudness in 2000 and have since both hit and non-hit song volume have began to dip back downward. Will this trend continue or will we all go deaf? Give it 10 more years to say for sure.
http://www.sciencedaily.com/releases/2011/12/111216234655.htm">ScienceDaily - Can Science Predict a Hit Song?
Photo Credit: Christopher Macsurak, www.chicagofabulousblog.com